12 citations
- F. Bach2 profiles21 · h 108
- Alessandro Rudi3 profiles9 · h 27
- Francis R. Bach4 profiles6 · h 7
- Francis Bach3 profiles5 · h 3
- P. Gaillard5 · h 25
- Adrien B. Taylor2 profiles3 · h 24
- Eliot Beyler3 · h 2
- Lawrence Stewart3 · h 6
- Pierre Gaillard2 profiles3 · h 2
- Blake E. Woodworth2 · h 21
- Claire Boyer2 · h 6
- Gérard Biau2 · h 6
- Département d'InformatiqueFR10 papers
- École Normale Supérieure - PSLFR5 papers
- Institut national de recherche en sciences et technologies du numériqueFR4 papers
- CERMICSFR2 papers
- International University of Fundamental StudiesRU2 papers
- Amazon (United States)US1 paper
- Bocconi UniversityIT1 paper
- Carnegie Mellon UniversityUS1 paper
- Centre de Robotique1 paper
- Centre National de la Recherche ScientifiqueFR1 paper
- Centrum Wiskunde & InformaticaNL1 paper
- College of Western IdahoUS1 paper
22 papers · 1 filter
Regularized Variational and Spectral Log-Density-Ratio Estimation in the Gaussian Location Model
Francis Bach
We study ridge-regularized log-density-ratio estimation in the Gaussian location model with a common covariance matrix. By affine invariance, the model is written as q N(0,…
A Spectral Framework for Closed-Form Relative Density Estimation
Francis Bach
We propose a closed-form spectral framework for relative log-density estimation in linearly parameterized probabilistic models, including unnormalized and conditional models. This…
A Convex Loss Function for Set Prediction with Optimal Trade-offs Between Size and Conditional Coverage
Francis Bach
We consider supervised learning problems in which set predictions provide explicit uncertainty estimates. Using Choquet integrals (a.k.a. Lov{á}sz extensions), we propose a convex…
Convergence of Deterministic and Stochastic Diffusion-Model Samplers: A Simple Analysis in Wasserstein Distance
Eliot Beyler, Francis Bach
We provide new convergence guarantees in Wasserstein distance for diffusion-based generative models, covering both stochastic (DDPM-like) and deterministic (DDIM-like) sampling met…
On the Effectiveness of the z-Transform Method in Quadratic Optimization
Francis Bach
The z-transform of a sequence is a classical tool used within signal processing, control theory, computer science, and electrical engineering. It allows for studying sequences from…
Optimal Denoising in Score-Based Generative Models: The Role of Data Regularity
Eliot Beyler, Francis Bach
Score-based generative models achieve state-of-the-art sampling performance by denoising a distribution perturbed by Gaussian noise. In this paper, we focus on a single determinist…